1
Assessmen o he pho o ol aic po en ial a u ban le el based
on 3D ci y models: A case s udy and new me hodological
app oach
Lau a Rome o Rod íguez P
a,*
P, E ic DuminilP
b
P, José Sánchez RamosP
a
P, U sula Eicke P
b
P.
P
a
P G upo de Te mo ecnia, Escuela Supe io de Ingenie os, Uni e sidad de Se illa. Camino de los
Descub imien os S/N, 41092 Se illa, Spain.
P
b
PResea ch Cen e o Sus ainable Ene gy Technologies, S u ga Uni e si y o Applied
Sciences. Schellings . 24, 70174 S u ga , Ge many.
* Co esponding au ho : Lau a Rome o Rod íguez. E-mail: 12TU[email p o ec ed]
ABSTRACT
The use o 3D ci y models combined wi h simula ion unc ionali ies allows o quan i y ene gy
demand and enewable gene a ion o a e y la ge se o buildings. The scope o his pape is
o de e mine he sola pho o ol aic po en ial a an u ban and egional scale using Ci yGML
geome y desc ip ions o e e y building. An inno a i e u ban simula ion pla o m is used o
calcula e he PV po en ial o he Ludwigsbu g Coun y in sou h-wes Ge many, in which e e y
building was simula ed by using 3D ci y models.
Bo h echnical and economic po en ial (conside ing oo a ea and insola ion h esholds) a e
in es iga ed, as well as wo di e en PV e iciency scena ios. In his way, i was possible o
de e mine he ac ion o he elec ici y demand ha can be co e ed in each municipali y and
he whole egion, deciding he bes s a egy, he p o i abili y o he in es men s and
de e mining op imal loca ions. Addi ionally, ano he impo an con ibu ion is a li e a u e
e iew ega ding he di e en me hods o PV po en ial es ima ion and he a ailable oo a ea
educ ion coe icien s. An economic analysis and emission assessmen has also been
de eloped.
The esul s o he s udy show ha i is possible o achie e high annual a es o co e ed
elec ici y demand in se e al municipali ies o some o he conside ed scena ios, eaching
e en mo e han 100% in some cases. The use o all a ailable oo space ( echnical po en ial)
could co e 77 % o he egion’s elec ici y consump ion and 56% as an economic po en ial
wi h only high i adiance oo s conside ed. The p oposed me hodological app oach should
con ibu e aluably in helping policy-making p ocesses and communica ing he ad an ages o
dis ibu ed gene a ion and PV sys ems in buildings o egula o s, esea che s and he gene al
public.
UKeywo ds
U ban ene gy consump ion; PV po en ial; U ban sola po en ial; Roo - op pho o ol aic
sys ems; Dis ibu ed Gene a ion; 3D ci y models.
2
1. In oduc ion
I is an undeniable ac ha ou p esen li ing s anda d s ongly depends on elec ici y and
o he o ms o ene gy. U baniza ion has led o a high inc ease in ene gy use, wi h buildings
being one o i s la ges con ibu o s and playing a signi ican ole on clima e change. As pa o
he sus ainabili y s a egy in Eu ope, he Ene gy Pe o mance o Buildings Di ec i e (EU, 2010)
and o he s such as he Renewable Ene gy Di ec i e (EU, 2009) ha e de ined a package o
measu es ha se s he pa h o no able and long e m imp o emen s in he ene gy
pe o mance o Eu ope׳s building s ock. Some examples a e he in oduc ion o Nea ly Ze o
Ene gy Buildings (NZEB) o he obliga ion o u ilize on-si e enewable ene gy. In addi ion, he
endency o new egula ions is o ex end he sys em bounda ies om a single building o he
u ban a ea, allowing he in e ac ion be ween di e en ene gy lows.
The new concep o dis ibu ed ene gy gene a ion is becoming inc easingly impo an , wi h
he e ec ha he dis ibu ion ne wo k is e ol ing om a once passi e powe -consuming o an
ac i e powe -gene a ing pa o he elec ic powe sys em (S ećko ić e al., 2016). Among he
di e en widesp ead dis ibu ed ene gy applica ions, he e is a g owing consensus ha he
deploymen o pho o ol aic (PV) sys ems in buildings is an a ac i e op ion. Analyses ha e
shown ha abou 60% o he oo a ea in Eu ope is sui able o sola echnologies (IEA, 2002;
Weiss e al., 2010), which could be sola he mal (SRTHR) o pho o ol aics. In his wo k he ocus
is on sola pho o ol aics. Howe e , in spi e o he ac ha he ad an ages o indi idual
buildings ha e been s udied, he e is li le unde s anding o he po en ial bene i s o an u ban
scale implemen a ion o such sys ems (Jo and O anica , 2011).
Elec ici y p oduc ion by PV is g owing wo ld-wide and g id-pa i y is a eali y in many places,
e en in low i adiance coun ies such as Sweden (Molin e al., 2016). Sola adia ion is a clean
and abundan sou ce o ene gy and PV is expec ed o con ibu e e en mo e signi ican ly in he
u u e, since oo ops p o ide la ge a eas sui able o sola ene gy exploi a ion. Howe e ,
unlike he non-u ban en i onmen wi h li le cons ain s o ene gy p oduc ion, buildings ha e
limi a ions on he a ailable a ea, and many ac o s ha e o be conside ed such as cons uc ion
es ic ions o obs uc ions due o he su oundings.
The be e he knowledge abou he PV po en ial and in es men cos o a egion, he easie i
is o help policy-making p ocesses, p e en u u e dispa i ies be ween supply and demand,
and communica e he ad an ages o building in eg a ed sys ems o he gene al public (F ei as
e al., 2015). The e o e, he i s s ep o his app oach is an analysis o de e mine he sola
po en ial o egions, which migh be a challenging ask due o he complexi y o he u ban
en i onmen .
Al hough a lo o esea ch has been p esen ed o measu e he PV po en ial o buildings and
plen y o s udies ha e ocused on he imp o emen o sola assessmen by de eloping
so wa e and algo i hms, 3D ci y models ha e no been made a ailable in public domain on a
ull-scale ye . In o de o es ima e he PV po en ial, di e en app oaches a e applied, om
simple es ima ions o ai bo ne LiDAR (Ligh De ec ion and Ranging) echnologies (Ho á h e
al., 2016). Depending on he scale and he le el o de ail equi ed, some me hodologies will be
mo e app op ia e han o he s.
3
In he las decade Ge many has expe ienced a massi e inc ease in cons uc ed PV plan s.
Howe e , only a small ac ion o he ins alled capaci y is in eg a ed wi hin buildings (S zalka
e al., 2012). The e is a la ge dispa i y be ween egions, which mo i a es in es iga ions o
egional po en ials ha acco ding o Mainze e al. (2014) ha e no been done in ea lie
epo s.
1.1. Aims and objec i es
The scope o his pape is o de e mine he PV po en ial a an u ban scale, which migh be
highly bene icial o u ban ene gy managemen conside ing di e en COR2R sa ing and
in es men app oaches. Bo h he echnical and economic po en ial a e in es iga ed and
iden i ied o each single building o he egion acco ding o i s speci ic oo shape ecei ing
sola adia ion.
The p esen s udy in oduces an inno a i e ool o he de e mina ion o he PV po en ial a an
u ban and egional scale by using 3D ci y models: he Ja a-based SimS ad pla o m (SimS ad ,
2016), which con ains simula ion models om he IN eg a ed Simula ion En i onmen
Language (INSEL, 2014), bo h de eloped a he S u ga Uni e si y o Applied Sciences. In
addi ion, a li e a u e e iew ega ding he di e en me hods o PV po en ial es ima ion and
a ailable oo a ea educ ion coe icien s has been ca ied ou .
Wi h he iew o showing i s ull capabili ies when dealing wi h PV po en ial analysis o whole
egions, SimS ad has been used in his s udy o es ima e he PV po en ial o he Ludwigsbu g
Coun y in sou h-wes Ge many (s a e o Baden-Wü embe g), in which e e y indi idual
building was simula ed (157724 buildings in o al). The main pu pose o his s udy is o
de e mine wha ac ion o he elec ici y demand can be co e ed in bo h each municipali y
and he whole egion, deciding he bes s a egy so as o each ha aim, he p o i abili y o
such in es men s and de e mining he op imal loca ions. An economic analysis and emission
assessmen has also been de eloped, as well as some insigh s in o he unce ain y o he PV
po en ial es ima ions.
2. Li e a u e e iew
2.1. Re iew o me hods o es ima ing he sola po en ial
The li e a u e e iew which has been ca ied ou shows ha he e a e many di e en
me hodologies which aim o de e mine he PV po en ial o a egion, bu as ye ew me hods
o assessing u ban scale impac s o sola ene gy sys em applica ions ha e been de eloped (Jo
and O anica , 2011). One o he mos impo an aspec s which should be bo ne in mind is he
scale, since he same echniques canno be applied a local, egional o con inen al le el.
Addi ionally, i is necessa y o know which da a is a ailable. Unlike Building In o ma ion Model
(BIM) s anda ds which se e as exchange suppo be ween di e en building ools allowing
high in e ope abili y, no comp ehensi ely applicable model s anda d exis s un il now o
U ban Ene gy Modelling (Nou el e al., 2015a). Tha is he eason why de elope s had o s a
om he beginning and c ea e hei own da a models.
As i has been men ioned be o e, he e a e many di e en app oaches when dealing wi h sola
po en ial es ima ions. The s udy pe o med by Schallenbe g-Rod íguez (2013) does a e y
comple e me hodology e iew and in e compa ison. Acco ding o i , he main di e ence
among he di e en p ocedu es is he me hod used o de e mine he oo a ea: based on he
a io oo su ace pe capi a, es ablishing a co ela ion be ween he popula ion densi y and he
oo a ea, o compu ing he o al oo a ea o he a ge egion. In (Li e al., 2015) he sola
4
po en ial in u ban esiden ial buildings is in es iga ed a di e en le els o si e densi ies,
compa ing he sola po en ial unde di e en u ban o ms whose o al a ailable oo a ea was
calcula ed wi h sample u ban se ings and wea he da a as he inpu s. O he possible op ions
a e based on building ypology h ough on-si e da a collec ion and isual inspec ion o a
ce ain a ea (Ho á h e al., 2016) o s a is ical calcula ion models which compu e he o al
oo a ea h ough ae ial objec -speci ic image ecogni ion (Ka e is e al., 2013). Ne e heless,
al hough he men ioned s udies include e y e icien and obus es ima ion models, hey
migh no be eplicable o he scope o he p esen s udy.
On he o he hand, he h ee mos impo an oo -a ea es ima ion me hods acco ding o
Melius e al. (2013) a e he ollowing:
-Cons an - alue me hods: hey a e a use ul s a ing poin o hei speed, bu hey make e y
simpli ied ule-o - humb assump ions such as he a io o il ed e sus la oo s, he numbe
o buildings wi h desi able oo op o ien a ions, o he amoun o space obs uc ed by building
componen s. The cons an s a e hen applied o he o al building s ock, de e mined om he
Census o example.
-Manual selec ion me hods: oo ops wi h cha ac e is ics ha appea sui able o PV a e
manually selec ed om sou ces such as ae ial pho og aphs and isually inspec ed o shading
and building obs uc ions. They a e mo e accu a e, bu e y ime-in ensi e and no easily
eplicable.
-Geog aphic In o ma ion Sys ems (GIS)-based me hods: used by he majo i y o analyses, hey
mainly use 3D models in o de o de e mine he a ailable oo op a ea o a egion, iden i y
obs uc ions o assess shadow e ec s on buildings. They a e much mo e accu a e and
eplicable bu compu e - esou ce in ensi e.
Ou ocus will be on GIS-based me hods, since hey can play a e y impo an pa in
suppo ing decision making by ackling he u gen ly equi ed ene gy ansi ion (Rami ez
Cama go e al., 2015). Fo e y p ecise calcula ions he mos app op ia e op ion is 3D
modeling and building simula ion (Ho á h e al., 2016). Ne e heless, his me hodology migh
only be applied o small-scale egions such as a ci y o a coun y due o he ac ha i is a ime-
consuming and esou ce-in ensi e p ocess (Ku dgelash ili e al., 2016).
3D ci y models ha e shown huge po en ials in he ield o ci y planning, and he numbe o
ci ies ep esen ed is inc easing exponen ially, a he same ime ha he in es men cos s and
ime equi ed o build hese models is dec easing hanks o new da a collec ion echnologies
such as LiDAR. D ones ha e also become a e y e icien and low-cos solu ion. An example o
s udy which makes use o 3D ci y models is he one p esen ed by Singh and Bane jee (2015),
which uses high-g anula i y land use da a a ailable in he public domain and GIS-based image
analysis o sa elli e images. Con e sely, Lukač e al. (2014) p esen a no el PV po en ial
es ima ion o e LiDAR da a, aking in o accoun he nonlinea e iciency cha ac e is ics o he
PV modules and in e e .
O he publica ions conside he ime se ies analysis o supply and demand (Rami ez Cama go
e al., 2015), assess he ime-dependen annual elec ical ene gy losses (S ećko ić e al., 2016),
build a Digi al Su ace Model (DSM) om LiDAR da a (Redweik e al., 2013), use o ho-image y
h ough cadas al da a (Be gamasco and Asina i, 2011a) o do objec o ien ed image analysis
and GIS combined wi h emo e sensing image da a o quan i y he a ailable oo a ea (Jo and
O anica , 2011). I should also be men ioned ha he use o PV can help mi iga e blackou
5
p oblems and assess he easibili y o oo op PV in emo e u ban a eas (Gau am e al., 2015).
A e y ho ough and aluable GIS-based s udy was de eloped by Mainze e al. (2014) o all
municipali ies in Ge many. Howe e , he s a is ical da a was assumed o be homogeneous (no
a ia ion in ypical building sizes be ween di e en municipali ies o example), apa om he
ac ha he PV po en ial o non- esiden ial buildings could no be assessed.
In o he s udies such as (S ećko ić e al., 2016) o (Khan and A salan, 2016), e en i GIS we e
used i was mainly o calcula e oo a eas, bu no o compu e sola p oduc ion, which will be
done in he p esen s udy. The e o e, he ou comes o his esea ch should con ibu e aluably
o he body o knowledge, since e y ew s udies ha e used bo h de ailed building and sola
i adiance da a o compu e he PV p oduc ion on speci ic si es (Schallenbe g-Rod íguez, 2013).
2.2. P ocess o de e mina ion o he a ailable oo a ea
Once he 3D model o he egion has been ob ained, i is possible o know he o al buil a ea
and he geome y o he buildings ha shape i . Howe e , many ci cums ances may lead o he
educ ion o he ini ial oo a ea. An ex ensi e li e a u e e iew has shown he g ea a ie y o
di e en educ ion coe icien s used o calcula ing he a ailable oo a ea o a egion. Mos
s udies ocus on he de e mina ion o oo and acade a eas, dis inguishing be ween la and
il ed oo s (Ku dgelash ili e al., 2016; Mainze e al., 2014; Melius e al., 2013; Schallenbe g-
Rod íguez, 2013) o be ween building ypes (Be gamasco and Asina i, 2011b; Schallenbe g-
Rod íguez, 2013).
The e seems o be an ag eemen so as o di e en ia e be ween a chi ec u al sui abili y and
sola sui abili y (By ne e al., 2015; Schallenbe g-Rod íguez, 2013). Howe e , hese s udies
di e since hei le el o de ail a ies, and he e is no common classi ica ion o hei
coe icien s. Some o hem gi e disagg ega ed ac o s (Be gamasco and Asina i, 2011b; By ne
e al., 2015; Izquie do e al., 2008; Schallenbe g-Rod íguez, 2013), while o he s show mo e
global ones (IEA, 2002; Mainze e al., 2014; Melius e al., 2013). In addi ion, unlike he
publica ions made by By ne e al. (2015) and Luque and Hegedus (2011) mos o hese s udies
do no conside he coe icien s o he sepa a ion o he PV panels (GCR) o he Se ice A ea
(SA), necessa y o main enance ope a ions.
A e ga he ing all he in o ma ion om ela ed s udies, i was decided o use o his s udy he
app oach shown in he lowcha in Figu e 1, which illus a es he way o calcula e he
u iliza ion ac o (UF). I should be no ed ha no p e ious s udy has used all o hese ac o s a
he same ime, bu only pa ially. This app oach includes all he educ ion coe icien s which
we conside as essen ial o ou s udy and shows he calcula ion p ocess o es ima ing he
a ailable oo a ea o PV pu poses, a e which calcula ions o he PV po en ial can be
pe o med.
Unlike p e ious publica ions in which hese coe icien s a e applied o he agg ega ed esul s
o a whole egion, ou s udy conside s hei applica ion o each building indi idually, which
inc eases he accu acy o he p ocedu e. This is due o he ac ha he 3D model allows us o
know hei cha ac e is ics, enabling us o apply di e en ac o s depending on he building
ha is being analyzed.

6
Figu e 1: Flowcha o he a ailable oo a ea calcula ion p ocess.
Wi h a iew o unde s anding he scope o each educ ion coe icien , hey a e going o be
b ie ly explained:
-Cons uc ion es ic ions (CRCONR): i e e s o space al eady occupied by elemen s loca ed on
he oo , such as ele a o s, ai ex ac o s, chimneys, s ai wells, wa e anks, HVAC ins alla ions
o windows.
-P o ec ed buildings (CRPROTR): his coe icien may be applied o buildings whe e o some eason
no acili y can be buil on, due o his o ical conside a ions o example.
-Shading e ec s (CRSHR): i conside s he shadowing p oduced by he oo i sel o by o he
buildings.
-Se ice A ea (CRSAR): necessa y space o main enance and access. A highe il angles he space
eed up due o he spacing be ween he PV panels (CRGCRR) can be used (By ne e al., 2015).
-O ien a ion losses (CRAZR): i akes in o accoun he ela i e amoun o sola adia ion which
eaches he su ace due o i s azimu h.
-Slope o he oo (CRSLR): i akes in o accoun he ela i e amoun o sola adia ion which
eaches he su ace due o he slope o he oo .
-Sepa a ion o he PV panels (CRGCRR): i conside s he dis ance be ween he panels so as o a oid
ecip ocal shadowing. Acco ding o Luque and Hegedus (2011), shade on as li le as 5-10% o
an a ay can educe i s ou pu by o e 80%.
-Ra io o PV panels (CRPVR): Ra io o he a ailable oo a ea used o ins all PV panels.
7
-Ra io o SRTHR panels (CRSTR): Ra io o he a ailable oo a ea used o ins all SRTHR panels.
2.3. P ocess o de e mina ion o he echnical PV po en ial
The PV po en ial is calcula ed in he way shown in Figu e 2.
Figu e 2: Flowcha o he echnical PV po en ial calcula ion p ocess.
-PV a ea (SRPVR): o al a ailable oo a ea used o ins all he PV panels [mP
2
P], de e mined by he
SimS ad so wa e (SimS ad , 2016).
-Incoming sola ene gy (IRPVR): annual insola ion in he PV modules su ace [kWh/mP
2
P•yea ] also
calcula ed by SimS ad .
-PV modules e iciency (ηRe R): e iciency o he PV modules depending on he echnology used.
-Tempe a u e and i adiance losses (ηRTHR): e iciency loss due o clima e cha ac e is ics. This
pa ame e is cu en ly objec o g ea in e es in he echnical communi y (Be gamasco and
Asina i, 2011b).
-Losses o o ien a ion (ηRAZR): i akes in o accoun he e lec ion losses due o non-no mal
incidence angle o he Sun’s ays (Li e al., 2015).
-Pe o mance a io (ηRPRR): losses due o con e sion e iciency o he in e e , cabling losses,
dus on he panels and o he s. Elec ici y s o age will no be conside ed in he p esen s udy.
3. Inpu da a and simula ion ools
3.1. Da a model and wea he p ocesso
Fo he modelling o he 3D building da a, he Open Geospa ial Conso ium (OGC) S anda d
Ci yGML (Ci yGML, 2012) has been chosen. Ci yGML is an open, mul i unc ional XML-based
da a model, a lexible spa io-seman ic da a o ma which o e s powe ul me hods o he
e alua ion o a ious analyses o ci y dis ic s, whole ci ies o egions.
A conside able ad an age o Ci yGML in compa ison wi h o he 3D ci y model o ma s is ha i
speci ies objec modelling in ou inc easing Le els o De ail (LOD1, LOD2, LOD3 and LOD4),
enabling he ci y model o adap o local building pa ame e a ailabili y. The mos simple
building ep esen a ion is LOD1, consis ing in a ec angula block. LOD2 includes he ull
building geome y wi h a ying heigh s o building pa s and he oo shape, LOD3 a de ailed
açade geome y including doo s and windows, and LOD4 he inclusion o indoo spaces. In
8
2014, he comple e building s ock o Ge many was modelled wi h Ci yGML – LOD1, and some
egions like Baden Wü embe g o Saxony ha e al eady comple ed hei 3D ci y model wi h
LOD2 (Nou el e al., 2015b). In o de o gene a e he 3D ci y models, LiDAR, s e eo ai pho o
o digi al cadas e enhanced wi h building in o ma ion can be used. In pa icula , lase
scanning me hods which a e o en used nowadays allow an au oma ic gene a ion o Ci yGML
models o whole ci ies in a sho ime.
On he o he hand, analyzing he sola po en ial o a egion equi es local wea he da a, ei he
hou ly o mon hly, in o de o know he ho izon al and di use adia ions, ambien
empe a u es, e c. The quali y o he sola adia ion da a depends on he sou ce, including
g ound s a ion measu emen s, sa elli e images o combina ions o bo h ypes (Assouline e al.,
2017).These da a a e impo ed in o he SimS ad pla o m h ough a wea he p ocesso om
di e en da abases such as PVGIS (PVGIS, 2012), INSEL (INSEL, 2014), o by using Me eono m
wea he iles chosen by he use .
3.2. U ban modeling pla o m SimS ad
Recen ly, u ban simula ion and 3D GIS ha e p og essed conside ably, bu wi hou no able
in e ac ion be ween hem. Wi h he pu pose o aking bo h domains in o accoun and
suppo ing public au ho i ies and enginee ing companies in he planning o he ene gy
ansi ion a u ban scale, he u ban ene gy simula ion pla o m SimS ad (SimS ad , 2016) was
de eloped by he S u ga Uni e si y o Applied Sciences in he amewo k o a p ojec unded
by he Ge man ede al Minis y o Economic A ai s and Ene gy.
Based on he open 3D Ci yGML models, i s wo k low-d i en s uc u e is highly modula and
ex ensible, allowing o a po en ially unlimi ed a ie y o u ban analysis p o ided ha he
equi ed da a is a ailable in he 3D model. Each wo k low s ep has hypo heses, pa ame e s
and in e media e esul s which can be modi ied and assessed h ough he G aphical Use
In e ace (GUI), enabling he use o c ea e scena ios acco dingly. In addi ion, i some
in o ma ion is no deducible o a ailable a building le el, such as building age necessa y o
hea demand calcula ions, de aul da a a e used om he building lib a y. In he case o PV
po en ial calcula ions all he equi ed in o ma ion is con ained in he Ci yGML model, as only
geome y da a a e used o he modeling.
The s a o he wo k low in SimS ad is he i ual 3D Ci yGML model. I should be no ed ha
SimS ad handles all LODs. Gi en he di e si y o he quali y o he 3D models, he nex s ep
would be he use o he healing module “Ci yDoc o ”, equi ed o check and co ec he
geome y o he model. Then, he da a-p ocessing allows o he comple ion o he model.
A e ha , he ene gy simula ions can be ca ied ou . SimS ad has he capaci y o ob aining
hou ly o mon hly da a in e e y simula ion, al hough he esul s o he p esen s udy a e
annual gi en ha he main goal is o es ima e he annual PV po en ial o a egion. The la es
e sion o SimS ad can pe o m a a ie y o mul i-scale ene gy analyses such as
hea ing/cooling demand diagnosis, building e u bishmen scena ios o pho o ol aic po en ial.
O he wo k lows a e unde way. Las o all, he esul s can be isualized in di e en ways wi h
pe o mance indices, g aphs o maps, as well as being expo ed o a ile.
9
Figu e 3: Example o adia ion map simula ed using SimS ad o de e mine op imal PV
loca ions in a municipali y.
Once he wea he da a a e a ailable, he adia ion p ocesso can compu e he incoming
i adiance on e e y building bounda y su ace, based on hei geome y and he di ec , di use
and ho izon al i adiances deli e ed by he wea he p ocesso .
In he cu en e sion o he SimS ad pla o m, he use can selec wo di e en adia ion
dis ibu ion models:
-INSEL model: based on he Hay sky model o di use i adiance calcula ion, equi es INSEL and
simula es sola i adiance on a bi a y su ace o ien a ions. I s execu ion ime is as and does
no depend on he 3D model size. Shading and in e - e lec ions a e no conside ed.
-Simpli ied Radiosi y Algo i hm (SRA): i is coupled wi h he Pe ez sky model, and conside s
bo h shadowing and he e lec ion e ec s o he su ounding buildings. I s execu ion ime
depends on he 3D model size and he amoun o simula ed buildings.
The cu en s udy will be based on he INSEL model wi hou shading due o he la ge amoun
o buildings ha will be analyzed. Shadowing e ec s will be app oxima ed h ough a educ ion
coe icien (see sec ion 5.1).
3.3. PV Po en ial analysis ool
Rega ding he PV po en ial ool included wi hin SimS ad , he sequen ial wo k low s eps a e
shown in Figu e 4. The inpu is he Ci yGML ile o he egion. Al hough i can also wo k wi h
LOD1, LOD2 is p e e able. LOD3 and LOD4 include mo e in o ma ion and hey could be
in e es ing o analyze acades o example, bu o he da a would be i ele an o ou pu pose
(such as in e nal pa i ions).
The ou pu s o his ool a e: i adiance, sui able oo a ea, nominal powe and annual ene gy
yield o e e y indi idual building. I is also able o pe o m he o e all calcula ions, as well as
show aluable g aphs and 3D maps wi h da a such as PV sui abili y in o de o assess op imal
loca ions.
16
Figu e 8: Pe cen age o he elec ici y demand co e ed in each municipali y by PV o he
economic po en ial s a egy.
Rega ding he agg ega ed alues o he whole coun y, he esul s can be seen in Figu e 9. The
main cha ac e is ics o he egion a e summa ized in Table 3.
Figu e 9: Pe cen age o he elec ici y demand co e ed by PV in he whole egion o he wo
scena ios: echnical and economic po en ial.
Va iable
Resul
To al elec ici y demand o he egion
1717 [GWh/yea ]
To al popula ion o he egion
354551 inhabi an s
To al numbe o buildings simula ed in SimS ad
157724 buildings
To al oo a ea o he egion
22.26 [km
P
2
P
]
To al a ailable oo a ea o he egion
11.14 [km
P
2
P
]
A e age % o la oo s:
16 %
A e age % o il ed oo s:
84 %
A e age su ace o olume a io o he buildings:
0.84 [m
P
-1
P
]
Table 3: Summa y o he cha ac e is ics o he Ludwigsbu g Coun y.
I PV modules could be ins alled on all he a ailable su ace ( echnical po en ial), using wa e -
based silicon modules (scena io A) could co e 77 % o he elec ici y demand o he egion.

17
On he o he hand, i hin- ilm modules we e used (scena io B wi h less e iciency), hen only
51% could be achie ed. I should be no ed ha he e iciency o he PV modules imp o es
e e y yea , so hese pe cen ages would inc ease acco dingly.
Con e sely, aking he economic po en ial in o accoun would esul in lowe payback pe iods,
bu he me elec ici y demand would be lowe han ha o he echnical po en ial. Wa e -
based silicon modules would co e 56% o he elec ici y demand, while hin- ilm based
modules would co e only 37 %. The summa y o he ob ained esul s o he egion is shown
in Table 4.
Scena io Va iable Resul
Desc ip ion o he a iable
calcula ed by SimS ad
RSCENARIO A
𝐸𝐸𝑃𝑃𝑃𝑃
𝑇𝑇𝑇𝑇𝑇𝑇ℎ𝑛𝑛𝑛𝑛𝑇𝑇𝑛𝑛𝑛𝑛
1318 [GWh/yea ]
Technical PV po en ial
𝑃𝑃𝑃𝑃𝑃𝑃
𝑇𝑇𝑇𝑇𝑇𝑇ℎ𝑛𝑛𝑛𝑛𝑇𝑇𝑛𝑛𝑛𝑛
1642 [MW
R
p
R
]
To al echnical PV nominal powe
𝐸𝐸𝑃𝑃𝑃𝑃
𝐸𝐸𝑇𝑇𝐸𝐸𝑛𝑛𝐸𝐸𝐸𝐸𝑛𝑛𝑇𝑇
957 [GWh/yea ]
Economic PV po en ial
𝑃𝑃𝑃𝑃𝑃𝑃
𝐸𝐸𝑇𝑇𝐸𝐸𝑛𝑛𝐸𝐸𝐸𝐸𝑛𝑛𝑇𝑇
1107 [MW
R
p
R
]
To al economic PV nominal powe
RSCENARIO B
𝐸𝐸𝑃𝑃𝑃𝑃
𝑇𝑇𝑇𝑇𝑇𝑇ℎ𝑛𝑛𝑛𝑛𝑇𝑇𝑛𝑛𝑛𝑛
872 [GWh/yea ]
Technical PV po en ial
𝑃𝑃𝑃𝑃𝑃𝑃
𝑇𝑇𝑇𝑇𝑇𝑇ℎ𝑛𝑛𝑛𝑛𝑇𝑇𝑛𝑛𝑛𝑛
1087 [MW
R
p
R
]
To al echnical PV nominal powe
𝐸𝐸𝑃𝑃𝑃𝑃
𝐸𝐸𝑇𝑇𝐸𝐸𝑛𝑛𝐸𝐸𝐸𝐸𝑛𝑛𝑇𝑇
644 [GWh/yea ]
Economic PV po en ial
𝑃𝑃𝑃𝑃𝑃𝑃
𝐸𝐸𝑇𝑇𝐸𝐸𝑛𝑛𝐸𝐸𝐸𝐸𝑛𝑛𝑇𝑇
744 [MW
R
p
R
]
To al economic PV nominal powe
Table 4: Summa y o he esul s ob ained by SimS ad o he wo di e en scena ios and
s a egies.
6.2. Emission calcula ions
Quan i ying he po en ial COR2R emission sa ings due o he implemen a ion o PV modules is
ano he impo an ou come ha may be in e ed om his s udy. This way, we a e able o
e alua e o each s a egy and scena io conside ed he amoun o COR2R emissions a oided and
he pe cen age o educ ion compa ed o he ini ial si ua ion, in which all he elec ici y is
ob ained om he g id.
Table 5 shows he alue o he COR2R emissions o he whole egion no ega ding any PV
sys ems.
Va iable
Resul
Desc ip ion
𝐶𝐶𝐶𝐶2,𝑡𝑡𝐸𝐸𝑡𝑡
918814 [ COR2R/yea ] Annual COR2R emissions.
𝐶𝐶𝐶𝐶2,𝑇𝑇𝐸𝐸𝑇𝑇𝑐𝑐
535 [gCOR2R/kWh]
Coe icien o CO
R2R
emissions in Ge many
(Umwel bundesam , 2016).
Table 5: COR2R emissions o R R he egion in he ini ial case.
To be e unde s and he emission sa ings, a ull Li e-Cycle Assessmen (LCA) would be
necessa y o e alua e he en i onmen al impac o he PV modules. Fo simplici y, a e
e iewing ela ed publica ions (Nugen and So acool, 2014; Peng e al., 2013) he p esen
s udy will conside a COR2R emission coe icien o 50 gCOR2R/kWh o he PV elec ici y
gene a ion.
The COR2R emissions a oided and he COR2 Remissions p oduced a e PV implemen a ion a e
calcula ed in he ollowing way:
𝐶𝐶𝐶𝐶2,𝑛𝑛𝑎𝑎𝐸𝐸𝑛𝑛𝑎𝑎𝑇𝑇𝑎𝑎 =𝐸𝐸𝑃𝑃𝑃𝑃 ∙(𝐶𝐶𝐶𝐶2,𝑇𝑇𝐸𝐸𝑇𝑇𝑐𝑐 −𝐶𝐶𝐶𝐶2,𝑇𝑇𝐸𝐸𝑇𝑇𝑐𝑐,𝑃𝑃𝑃𝑃)
𝐶𝐶𝐶𝐶2,𝑝𝑝𝑝𝑝𝐸𝐸𝑎𝑎𝑝𝑝𝑇𝑇𝑇𝑇𝑎𝑎 =𝐸𝐸𝑃𝑃𝑃𝑃 ∙𝐶𝐶𝐶𝐶2,𝑇𝑇𝐸𝐸𝑇𝑇𝑐𝑐,𝑃𝑃𝑃𝑃 + (𝐸𝐸𝐷𝐷𝑇𝑇𝐸𝐸𝑛𝑛𝑛𝑛𝑎𝑎 − 𝐸𝐸𝑃𝑃𝑃𝑃)∙𝐶𝐶𝐶𝐶2,𝑇𝑇𝐸𝐸𝑇𝑇𝑐𝑐
18
Table 6 p esen s he esul s o he echnical and economic po en ial o he wo conside ed
scena ios.
SCENARIO
S a egy
CO
R2R
emissions
a oided
[ CO
R
2
R
/yea ]
CO
R2R
emissions
p oduced
[ CO
R
2
R
/yea ]
COR2R sa ings
achie ed [%]
SCENARIO A
Technical
po en ial
639141
279674
70%
Economic
po en ial
464262
454552
51%
SCENARIO B
Technical
po en ial
422907
495907
46%
Economic
po en ial
312288
606527
34%
Table 6: COR2R emissions o R R he egion o each app oach o he s udy.
The esul s show he huge po en ial con ibu ion o oo op PV o he educ ion o he COR2R
emissions (and he e o e o he pollu an s).
6.3. Economic easibili y
Ano he pu pose o he p esen s udy was o de elop an economic analysis o he
implemen a ion o PV modules in he egion ega ding he p oposed s a egies, so as o assess
hei easibili y.
Fo he calcula ions, i is assumed ha 30% o he PV p oduc ion o he egion will be used o
sel -consump ion (IEA-PVPS, 2016), while he emaining 70 % will bene i om he eed-in-
a i s de ised by he go e nmen . In addi ion, i will be conside ed ha main enance o he
sys ems would annually incu addi ional cos s o 4% o he co esponding in es men .
The o al in es men cos s 𝐶𝐶𝑡𝑡 [€] we e es ima ed h ough he o al nominal ins alled powe
𝑃𝑃𝑃𝑃𝑃𝑃 [kWRpR], and he annual sa ings 𝐴𝐴𝑠𝑠 [€/yea ] (by a oiding he elec ici y cos s) we e
iden i ied and calcula ed in he ollowing way:
𝐶𝐶𝑡𝑡=𝑃𝑃𝑃𝑃𝑃𝑃 • 𝐶𝐶𝑆𝑆𝑆𝑆𝑠𝑠𝑡𝑡𝑇𝑇𝐸𝐸
𝐴𝐴𝑠𝑠=𝐸𝐸𝑃𝑃𝑃𝑃 • �𝐹𝐹𝑆𝑆𝑇𝑇𝑛𝑛𝑐𝑐 •𝐶𝐶𝑇𝑇𝑛𝑛𝑇𝑇𝑇𝑇 +�1− 𝐹𝐹𝑆𝑆𝑇𝑇𝑛𝑛𝑐𝑐�•𝐶𝐶𝑐𝑐𝑡𝑡�− 𝐶𝐶𝑡𝑡•𝐹𝐹
𝐸𝐸
The chosen ac o s which we e applied o he calcula ions a e shown in Table 7.
Va iable
Resul
Desc ip ion
𝐶𝐶𝑇𝑇𝑛𝑛𝑇𝑇𝑇𝑇
0.22 [€/kWh]
Elec ici y p ice pe kWh
(Expe ience alue).
𝐶𝐶𝑐𝑐𝑡𝑡
0.1231 [€/kWh]
Feed-in a i o small PV
acili ies in Ge many
(Bundesne zagen u , 2015).
𝐶𝐶
𝑆𝑆𝑆𝑆𝑠𝑠𝑡𝑡𝑇𝑇𝐸𝐸 1280 [€/kWp]
A e age p ice o he
ins alla ion o 1 kWp PV (ISE
F aunho e Ins i u e o Sola
Ene gy, 2016).
𝐹𝐹𝑆𝑆𝑇𝑇𝑛𝑛𝑐𝑐
30 [%]
Pe cen age o he elec ici y
used o sel -consump ion.
𝐹𝐹
𝐸𝐸
4 [%]
Annual pe cen age o
main enance cos s.
Table 7: Lis ing o economic indica o s and hei pa ame e s o he PV po en ial o he egion.
19
A e ex ac ing he equi ed a iables om SimS ad and applying he chosen economic
indica o s, he esul s we e ob ained o each p oposed s a egy and scena io (see Table 8).
SCENARIO
S a egy
Ene gy yield
[GWh/yea ]
Nominal
powe
[MWp]
To al
In es men
Cos s [M€]
To al
Annual
Sa ings
[M€/yea ]
𝐸𝐸𝑃𝑃𝑃𝑃
𝑃𝑃𝑃𝑃𝑃𝑃
𝐶𝐶𝑡𝑡
As
SCENARIO A
Technical po en ial
1318
1642
2101
116
Economic po en ial
957
1107
1416
89
SCENARIO B
Technical po en ial
872
1087
1391
77
Economic po en ial
644
744
953
60
Table 8: Economic esul s o each p oposed s a egy and scena io.
Se e al indings can be deduced om he p esen ed esul s. As can be seen, he economic
po en ial o bo h scena ios would ansla e in o much lowe necessa y in es men cos s o
he implemen a ion o he PV modules compa ed o he echnical po en ial s a egies.
Ne e heless, i would also mean less PV yield, annual elec ici y sa ings and emissions
educ ion.
The PV economic expec a ions could be enhanced h ough echnological inno a ions allowed
by economies o scale. This would make he p ojec s mo e p o i able, and a ac new
in es o s. The emo al o adminis a i e ba ie s by he go e nmen s hemsel es and
incen i es o pe suade he popula ion abou he use ulness o PV sys ems on buildings should
be emphasized in some coun ies, in o de o allow o a widesp ead implemen a ion o hese
p omising solu ions as a as sus ainable de elopmen and ene gy conse a ion a e conce ned.
6.4. Unce ain y o he me hod
Acknowledging he unce ain y o PV po en ial es ima ion me hods is a majo poin o u he
esea ch, since i is no equen ly p esen in many s udies. Limi ed inpu da a and he use o
de aul alues a e impo an sou ces o unce ain y, as well as simpli ica ions and hypo heses.
The a ia ions o sola adia ion also ha e o be aken in o accoun . Depending on he
ques ion, ei he long e m a e age wea he iles o wea he da a o a speci ic yea unde
conside a ion should be used.
In he case o PV po en ial es ima ions all he equi ed in o ma ion is geome y da a,
con ained in he 3D model. The highe he Le el o De ail, he mo e accu a e he PV
es ima ions. As an example, in o de o e alua e he a ia ions o using di e en LOD’s he
agg ega ed esul s o each municipali y ega ding he echnical PV po en ial ha e been
compa ed o wha he esul s would ha e been i modeled in LOD1, which conside s all oo s
o be la . The esul s a e shown in Figu e 10.
20
Figu e 10: Pe cen age o di e ence be ween he PV po en ial yield conside ing LOD2 and LOD1
models.
As is appa en , in e e y case a LOD1 model would unde es ima e he PV po en ial o he
egion. The explana ion lies in he ac ha al hough conside ing only la oo s wi h a sou h
acing il ed PV gene a o would mean highe speci ic adia ion, he module sepa a ion o
a oid shading would educe he use ul oo a ea by a ac o o 0.46, as well as o he educ ion
coe icien s which a e mo e es ic i e in la oo s. The po en ial oo a ea is always
unde es ima ed in he LOD1 model. The ins alled module a ea in he egion unde he
assump ion o only la oo s is be ween 6.67 % and 13.34 % lowe han o he LOD2 oo
s uc u e wi h p edominan ly il ed oo s. In ac , he wo municipali ies in Figu e 10 wi h a
highe pe cen age o di e ence a e he ones wi h a highe a e age il angle o hei oo s.
None heless, he di e ences a e a he small, so in case o ha ing a LOD1 model he esul s
can be expec ed o be accu a e enough.
I should also be no ed ha using educ ion coe icien s in o de o assess cons uc ion
es ic ions in oo s o he in luence o ees and buildings is ano he impo an sou ce o
unce ain y. An inc ease in he le el o de ail o he 3D models which includes his in o ma ion
could eplace in he u u e hese educ ion coe icien s wi h mo e accu a e alues o each
indi idual building.
Wi h ega ds o he alida ion o he esul s, esea che s ha e equen ly li le in o ma ion
abou he accu acy o hei es ima es (Melius e al., 2013). In o de o alida e ou esul s, he
ou comes ob ained by Mainze e al. ,2014 (whose s udy conside s he PV po en ial o all he
egions in Ge many) ha e been consul ed. In he Ludwigsbu g Coun y a ea, hey ob ained a
alue o echnical po en ial g ea e han 1000 MWh/kmP
2
P and 1000-4000 kWp/kmP
2
P o he
egion. The esul s o he echnical po en ial in ou s udy o Scena io A a e 1443.5 MWh/kmP
2
P
and 1799.4 kWp/kmP
2
P, and o Scena io B 955.8 MWh/kmP
2
P and 1191.0 kWp/kmP
2
P. The e o e,
he esul s a e qui e consis en wi h he ones ob ained by hem. As s a ed in (F ei as e al.,
2015), i is expec ed ha as u he and mo e sophis ica ed sola maps and u he and mo e
di e se ins alla ion case s udies a e published, an in e ac i e dialogue be ween hese wo
esea ch a eas will lead o model alida ion and imp o emen .
21
7. Conclusions
This pape p oposes o use 3D u ban da a models based on he Ci yGML s anda d o analyze
he pho o ol aic po en ial on an u ban and e en egional scale. The simula ion me hodology is
based on a building by building oo su ace analysis and i adiance simula ion and ca e ully
e ises educ ion ac o s o he ene gy yield de e mina ion, applying hem o each building
sepa a ely. Realis ic s a egies and scena ios o PV implemen a ion we e de eloped in a case
s udy egion in Ge many. Economic calcula ions ha e also been pe o med so as o analyze he
easibili y o he equi ed in es men s.
Acco ding o he esul s ob ained, i is possible o achie e high a es o elec ici y demand
co e ed by PV in many municipali ies (e en mo e han 100% o low densi y municipali ies,
which means an elec ici y su plus). Wi hin he en i e egion wi h 34 municipali ies
in es iga ed, PV sys ems could gene a e 77 % o he elec ici y consump ion by using all
a ailable oo space, p oducing a o al o 1318 GWh/yea h ough he ins alla ion o 1642
MWp, hus educing he CO2 emissions no iceably. Con e sely, 56% o he elec ici y demand
could be p oduced i only oo s wi h enough insola ion and a minimum su ace a ea o an
economically easible PV ins alla ion a e used. To ealize he economically iable PV
ins alla ions and educe he elec ici y ela ed CO2 emissions by 51%, he es ima ed
in es men pe capi a is a ound 4000 Eu os o a o al o 1416 million Eu os in he Coun y. In
conclusion, i p ope ly designed hese PV sys ems could signi ican ly dec ease p ima y ene gy
consump ion and emissions, ea i ming hei use ulness and he impo an ole hey can play
in he nea u u e.
8. Fu u e wo k
Du ing he de elopmen o his esea ch wo k, some u u e di ec ions ha e been iden i ied
which could esul in mo e p ecise PV po en ial calcula ions. Fi s o all, due o he la ge
amoun o buildings and he equi ed compu a ional ime o mo e sophis ica ed adia ion
p ocesso s, he Hay model was used in his s udy o analyze all he in ol ed municipali ies.
Since he in e ac ion be ween buildings was no aken in o accoun , a shadowing educ ion
ac o was conside ed, aken om he li e a u e e iew. Howe e , his educes he accu acy o
he p ocedu e which makes use o p ecise geome y building models. In he u u e, he use o
iling s a egies ha a e cu en ly unde de elopmen will educe he equi ed compu a ional
ime, consequen ly making he calcula ions easible o he SRA adia ion model, which
conside s he in luence be ween buildings and is al eady implemen ed wi hin SimS ad . This
imp o emen will also help o make he la ge scale mo e aluable.
Addi ionally, he p esen s udy has only conside ed oo su aces, bu i could be ex ended o
acades. The impo ance o u he esea ch ega ding he unce ain y o he PV po en ial
es ima ions should also be highligh ed. Las o all, he p og ess in he ield o 3D modelling will
ensu e in he u u e ha models wi h highe LODs a e a ailable, hus inc easing he accu acy
o he PV po en ial analyses.

22
URe e ences
Assouline, D., Mohaje i, N., Sca ezzini, J.-L., 2017. Quan i ying oo op pho o ol aic sola
ene gy po en ial: A machine lea ning app oach. Sol. Ene gy 141, 278–296.
doi:10.1016/j.solene .2016.11.045
Be gamasco, L., Asina i, P., 2011a. Scalable me hodology o he pho o ol aic sola ene gy
po en ial assessmen based on a ailable oo su ace a ea: Fu he imp o emen s by
o ho-image analysis and applica ion o Tu in (I aly). Sol. Ene gy 85, 2741–2756.
doi:10.1016/j.solene .2011.08.010
Be gamasco, L., Asina i, P., 2011b. Scalable me hodology o he pho o ol aic sola ene gy
po en ial assessmen based on a ailable oo su ace a ea: Applica ion o Piedmon
Region (I aly). Sol. Ene gy 85, 1041–1055. doi:10.1016/j.solene .2011.02.022
Bundesne zagen u , 2015. <h p://www.bundesne zagen u .de>.
By ne, J., Taminiau, J., Ku dgelash ili, L., Kim, K.N., 2015. A e iew o he sola ci y concep and
me hods o assess oo op sola elec ic po en ial, wi h an illus a i e applica ion o he
ci y o Seoul. Renew. Sus ain. Ene gy Re . 41, 830–844. doi:10.1016/j. se .2014.08.023
Ci yGML, 2012. Exchange and s o age o i ual 3D ci y models. <h p://www.ci ygml.o g/>.
Compagnon, R., 2004. Sola and dayligh a ailabili y in he u ban ab ic. Ene gy Build. 36, 321–
328. doi:10.1016/j.enbuild.2004.01.009
Eicke , U., Nou el, R., Duminil, E., Coo s, V., 2014. Assessing passi e and ac i e sola ene gy
esou ces in ci ies using 3D ci y models. Ene gy P ocedia 57, 896–905.
doi:10.1016/j.egyp o.2014.10.299
EU, 2010. Di ec i e 2010/31/EU o he Eu opean Pa liamen and o he Council o 19 May 2010
on he ene gy pe o mance o buildings ( ecas ). O . J. Eu . Union 13–35.
doi:10.3000/17252555.L_2010.153.eng
EU, 2009. Di ec i e 2009/28/EC o he Eu opean Pa liamen and o he Council o 23 Ap il
2009. O . J. Eu . Union 140, 16–62. doi:10.3000/17252555.L_2009.140.eng
F ei as, S., Ca i a, C., Redweik, P., B i o, M.C., 2015. Modelling sola po en ial in he u ban
en i onmen : S a e-o - he-a e iew. Renew. Sus ain. Ene gy Re . 41, 915–931.
doi:10.1016/j. se .2014.08.060
Gau am, B.R., Li, F., Ru, G., 2015. Assessmen o u ban oo op sola pho o ol aic po en ial o
sol e powe sho age p oblem in Nepal. Ene gy Build. 86, 735–744.
doi:10.1016/j.enbuild.2014.10.038
G een, M.A., Eme y, K., Hishikawa, Y., Wa a, W., Dunlop, E.D., 2016. Sola cell e iciency ables
( e sion 47). P og. Pho o ol aics Res. Appl. 24, 3–11. doi:10.1002/pip.2728
Ho á h, M., Kassai-Szoó, D., Csoknyai, T., 2016. Sola ene gy po en ial o oo s on u ban le el
based on building ypology. Ene gy Build. 111, 278–289.
doi:10.1016/j.enbuild.2015.11.031
IEA, 2002. Po en ial o building in eg a ed pho o ol aics. IEA-PVPS Task 2002, 2–4.
IEA-PVPS, 2016. Re iew and analysis o sel -consump ion policies. Repo IEA-PVPS T1-28:2016.
23
INSEL, 2014. A Simula ion Sys em o Renewable Ene gy Supply Sys ems. Ve sion 8.2.
<h p://www.insel.eu>.
ISE F aunho e Ins i u e o Sola Ene gy, 2016. Pho o ol aics epo .
Izquie do, S., Rod igues, M., Fueyo, N., 2008. A me hod o es ima ing he geog aphical
dis ibu ion o he a ailable oo su ace a ea o la ge-scale pho o ol aic ene gy-
po en ial e alua ions. Sol. Ene gy 82, 929–939. doi:10.1016/j.solene .2008.03.007
Jo, J.H., O anica , T.P., 2011. A hie a chical me hodology o he mesoscale assessmen o
building in eg a ed oo sola ene gy sys ems. Renew. Ene gy 36, 2992–3000.
doi:10.1016/j. enene.2011.03.038
Ka e is, M., Slini, T., Papadopoulos, A.M., 2013. U ban sola ene gy po en ial in G eece: A
s a is ical calcula ion model o sui able buil oo a eas o pho o ol aics. Ene gy Build.
62, 459–468. doi:10.1016/j.enbuild.2013.03.033
Khan, J., A salan, M.H., 2016. Es ima ion o oo op sola pho o ol aic po en ial using geo-
spa ial echniques: A pe spec i e om planned neighbo hood o Ka achi - Pakis an.
Renew. Ene gy 90, 188–203. doi:10.1016/j. enene.2015.12.058
Ku dgelash ili, L., Li, J., Shih, C.-H., A ia, B., 2016. Es ima ing echnical po en ial o oo op
pho o ol aics in Cali o nia, A izona and New Je sey. Renew. Ene gy 95, 286–302.
doi:10.1016/j. enene.2016.03.105
Li, D., Liu, G., Liao, S., 2015. Sola po en ial in u ban esiden ial buildings. Sol. Ene gy 111, 225–
235. doi:10.1016/j.solene .2014.10.045
Lukač, N., Seme, S., ZLaus, D., S umbe ge , G., Zalik, B., 2014. Buildings oo s pho o ol aic
po en ial assessmen based on LiDAR (Ligh De ec ion And Ranging) da a. Ene gy 66,
598–609. doi:10.1016/j.ene gy.2013.12.066
Luque, A., Hegedus, S., 2011. Handbook o Pho o ol aic Science and Enginee ing, Second. ed.
John Wiley and Sons. doi:10.1002/9780470974704
Mainze , K., Fa h, K., Mckenna, R., S engel, J., Fich ne , W., Schul mann, F., 2014. A high-
esolu ion de e mina ion o he echnical po en ial o esiden ial- oo -moun ed
pho o ol aic sys ems in Ge many. Sol. Ene gy 105, 715–731.
doi:10.1016/j.solene .2014.04.015
Melius, J., Ma golis, R., Ong, S., 2013. Es ima ing Roo op Sui abili y o PV : A Re iew o
Me hods , Pa en s , and Valida ion Techniques. Technical Repo NREL/TP-6A20-60593.
Mohaje i, N., Upadhyay, G., Gudmundsson, A., Assouline, D., Kämp , J., Sca ezzini, J.L., 2016.
E ec s o u ban compac ness on sola ene gy po en ial. Renew. Ene gy 93, 469–482.
doi:10.1016/j. enene.2016.02.053
Molin, A., Schneide , S., Rohdin, P., Mosh egh, B., 2016. Assessing a egional building applied
PV po en ial – Spa ial and dynamic analysis o supply and load ma ching. Renew. Ene gy
91, 261–274. doi:10.1016/j. enene.2016.01.084
Nou el, R., Kaden, R., Bahu, J., Kaemp , J., Cip iano, P., Laus e , M., Benne , J., Munoz, E.,
Tou nai e, O., Caspe , E., 2015a. Genesis o he Ci yGML Ene gy ADE. Cisba 2015 931–
936. doi:10.5075/ep l-cisba 2015-931-936
Nou el, R., Mas ucci, A., Leopold, U., Baume, O., Coo s, V., Eicke , U., 2015b. Combining GIS-
based s a is ical and enginee ing u ban hea consump ion models: Towa ds a new
amewo k o mul i-scale policy suppo . Ene gy Build. 107, 204–212.
24
doi:10.1016/j.enbuild.2015.08.021
Nugen , D., So acool, B.K., 2014. Assessing he li ecycle g eenhouse gas emissions om sola
PV and wind ene gy: A c i ical me a-su ey. Ene gy Policy 65, 229–244.
doi:10.1016/j.enpol.2013.10.048
Peng, J., Lu, L., Yang, H., 2013. Re iew on li e cycle assessmen o ene gy payback and
g eenhouse gas emission o sola pho o ol aic sys ems. Renew. Sus ain. Ene gy Re . 19,
255–274. doi:10.1016/j. se .2012.11.035
PVGIS, 2012. Pho o ol aic Geog aphical In o ma ion Sys em.
<h p:// e.j c.ec.eu opa.eu/p gis/>.
Rami ez Cama go, L., Zink, R., Do ne , W., S oeglehne , G., 2015. Spa io- empo al modeling o
oo - op pho o ol aic panels o imp o ed echnical po en ial assessmen and elec ici y
peak load o se ing a he municipal scale. Compu . En i on. U ban Sys . 52, 58–69.
doi:10.1016/j.compen u bsys.2015.03.002
Redweik, P., Ca i a, C., B i o, M., 2013. Sola ene gy po en ial on oo s and acades in an u ban
landscape. Sol. Ene gy 97, 332–341. doi:10.1016/j.solene .2013.08.036
Schallenbe g-Rod íguez, J., 2013. Pho o ol aic echno-economical po en ial on oo s in egions
and islands: The case o he Cana y Islands. Me hodological e iew and me hodology
p oposal. Renew. Sus ain. Ene gy Re . 20, 219–239. doi:10.1016/j. se .2012.11.078
SimS ad , 2016. <h p://www.sims ad .eu/en/index.h ml>.
Singh, R., Bane jee, R., 2015. Es ima ion o oo op sola pho o ol aic po en ial o a ci y. Sol.
Ene gy 115, 589–602. doi:10.1016/j.solene .2015.03.016
S ećko ić, N., Lukač, N., Žalik, B., Š umbe ge , G., 2016. De e mining oo su aces sui able o
he ins alla ion o PV (pho o ol aic) sys ems, based on LiDAR (Ligh De ec ion And
Ranging) da a, py anome e measu emen s, and dis ibu ion ne wo k con igu a ion.
Ene gy 96, 404–414. doi:10.1016/j.ene gy.2015.12.078
S zalka, A., Alam, N., Duminil, E., Coo s, V., Eicke , U., 2012. La ge scale in eg a ion o
pho o ol aics in ci ies. Appl. Ene gy 93, 413–421. doi:10.1016/j.apene gy.2011.12.033
Umwel bundesam , 2016. En wicklung de spezi ischen kohlendioxid-emissionen des
deu schen s ommix in den Jah en 1990 bis 2015.
Weiss, W., Bie may , P., 2010. Po en ial o Sola The mal in Eu ope, Repo o he EU- unded
p ojec RESTMAC.
Wigin on, L.K., Nguyen, H.T., Pea ce, J.M., 2010. Quan i ying oo op sola pho o ol aic
po en ial o egional enewable ene gy policy. Compu . En i on. U ban Sys . 34, 345–
357. doi:10.1016/j.compen u bsys.2010.01.001